152 ANNUAL OF SCIENTIFIC DISCOVERY. 



given by Angstrom, with some observations made some years 

 since by Fizean, shows a remarkable coincidence of results ob- 

 tained by different methods, and is a new confirmation of the 

 truth of the undulatory theory of light. 



Fizean. produced the phenomenon of Newton's Rings by laying 

 a convex lens of very long focus upon a piece of glass with plane 

 parallel surfaces, and illuminating the combination by the mono- 

 chromatic light of the soda flame. The lens was so arranged 

 that it could be left to touch the glass or could be separated from 

 it by a known distance, measured by a micrometer screw. On 

 separating the lens from the glass plate, the rings were seen to 

 move in towards the centre of the lens, where they successively 

 disappeared, while their place was supplied by fresh rings which 

 made their appearance at the circumference of the lens. Fizeau 

 found that when the phenomenon was observed with sufficient care 

 nearly 500 rings could be counted, flowing inwards one after an- 

 other, but that after about this number the rings ceased to be visible, 

 the surface of the glass showing a nearly uniform illumination 

 all over, instead of a sharply defined alternation of light and dark 

 bands. When, however, the distance between the lens and the 

 glass plate was further increased, the rings reappeared, getting 

 gradually more and more distinct, until when nearly another 500 

 had passed they had become as sharp as at first ; but a still further 

 increase of distance caused them again to become confused, and 

 they ceased a second time to be discernible at about the fifteen- 

 hundredth. With a still greater separation of the glasses, how- 

 ever, they reappeared again, and became quite sharp at the two- 

 thousandth, after which, -for a third time, they got gradually 

 confused, and became indistinguishable at about the twenty-five- 

 hundredth. So the phenomenon, went on as the glasses were 

 separated, and not until fifty-two such groups had been counted 

 did the bands finally cease to be distinguishable. The two glasses 

 were then separated by an interval of fifteen millimeters, or more 

 than half an inch. 



This remarkable phenomenon of the alternate periods of dis- 

 tinctness and confusion of the rings is easily explained, as M. 

 Fizeau points out, when we remember that the light employed 

 was not strictly homogeneous, but consisted of two portions of 

 nearly but not quite equal degrees of refrangibility. If either of 

 these two constituent parts of the light had been used by itself, it 

 would have produced a set of rings, but the rings of one set would 

 have been a very little larger than the corresponding rings of the 

 other. Hence, if the two sets of rings are put together (as they 

 were in Fizeau's experiments), they will nearly but not quite fit 

 each other. If we examine a few rings at the centre, when the 

 two glasses are in contact, they will appear to coincide precisely ; 

 but if they are traced to a sufficient distance from the centre the 

 coincidence is seen not to be exact. For although the twentieth 

 (say) of one set is not perceptibly bigger than the twentieth ring 

 of the other set, the five-hundredth of one set is perceptibly bigger 

 than the five-hundredth of the other, and when put upon it falls 

 almost exactly half way between the five-hundredth and five- 



